From: Reini U. <ru...@x-...> - 2005-01-14 19:02:31
|
Joel Uckelman schrieb: > I think I have a solution for bug 1090579, https://sourceforge.net/tracker/i > ndex.php?func=detail&aid=1090579&group_id=6121&atid=106121. I'd appreciate > it if someone more familiar with the markup parsing code than I am would > take a look at my suggested fix and tell me if it seems sane before I > commit it and close the bug. Your "__" patch to prevent from bold detection. This is indeed hairy code. Shouldn't this also be done for italic? Better would be to urlencode/decode those magic tags in pagenames/urls: "__", "''" *PageName is this a list or link to "*PageName" I would say a list entry. #PageName is this a list or link to "#PageName" I would say a list entry. Legal pagenames: This is unfortunately database backend dependent. Flatfile and cvs have some filesystem and security related restrictions. All other should have no restrictions at all. Currently problems exists with (out of my head) "0": related to php vbasic-like empty() and ! checks. already fixed in some cases, but there still some holes. "[" and "]" cause problems with the p[] arg in WikiAdmin plugins. I think I fixed this but I'm not sure) "<" ">" : forgot why. parser related probably. (i.e. [Page<em>Name]) utf-8 special chars are detected and fixed upfront. IE sends urls in utf8, fixTitleEncoding() -- Reini Urban http://xarch.tu-graz.ac.at/home/rurban/ |